In humans, Salmonella infections cause two major clinical diseases: salmonellosis and typhoid fever. Silent carriage of the bacteria is frequent and contributes to disease dissemination. Using a genomic approach, we have reported the identification of ten loci (Ses1-Ses10) affecting Salmonella persistence in mice. A major locus, Ses1, was validated using a congenic approach. Nramp1 remains a strong candidate gene for Ses1 although we did not detect a significant interaction between Ses1 and Nramp1-/-. We also present the creation of new double congenic strains (Ses1/Ses4 and Ses1/Ses5) that will be used to validate the inheritance model of Salmonella clearance in females. Furthermore, the influence of Nramp1 on the transcriptome of Salmonella was investigated and diverse virulence mechanisms were shown to be involved. Notably, differential phoP expression, and the resulting differential expression of PhoP-regulated genes, was observed in the presence of Nramp1. Our results confirm the importance of host-pathogen interactions in determining the outcome of infection. / Les infections à salmonelles regroupent différentes maladies dont la salmonellose et la fièvre typhoïde. Le portage asymptomatique des salmonelles est fréquent et contribue à la dissémination de la maladie. En utilisant une approche de criblage génomique par locus, nous avons identifié dix loci (Ses1-Ses10) affectant la persistance de Salmonella chez la souris. Un locus majeur, Ses1, a été validé en utilisant des souris congéniques. Nramp1 demeure un gène candidat de choix pour Ses1 quoiqu’un test d’interaction Ses1/Nramp1-/- se soit avéré non significatif. Le modèle proposé de portage de Salmonella incluant des interactions entre les loci Ses1/Ses4 et Ses1/Ses5 sera exploré par la création de nouvelles lignées congéniques combinatoires qui ont été créées durant la préparation de cette thèse. L'influence de Nramp1 sur le transcriptome de Salmonella a été étudiée au niveau des mécanismes bactériens de virulence. En particulier, nous avons observé une expression différentielle de phoP, et par conséquent l’expression différentielle de gène dont l’expression est contrôlée par PhoP, en présence de Nramp1. Nos résultats confirment l'importance des interactions hôte-pathogène dans l’issue de l'infection à salmonelles.

Molecular analysis of DNA variation has usurped the place of all earlier technologies in forensic identification of victims and suspects alike. Although the field of ethics has made attempts to cope with the plethora of available genetic information, especially in clinical application, there has been little scrutiny of emerging ethical issues in the forensic domain. Legal scholarship highlights some aspects of the emerging issues, with particular relevance to the challenges faced in court and those regarding individual liberties. The overall objective of this thesis was to evaluate the scientific validity, ethical acceptability and legal accountability of the forensic applications of molecular genetics. In particular, contemporary science has allowed us to access information far beyond what was originally anticipated, such that trace DNA can be obtained trivially from any individual. As a consequence, the scope and composition of existing DNA banks far exceeds the legislative mandate. Chapter 1 reviews the current legal standards for evidence and assesses the level of exactitude necessary for forensic DNA testing to meet evidentiary standards. An evaluation of current practices in DNA banking revealed adequate informed consent practices; the need for a re-examination of access to public health samples with attention to local population interests and the necessity for developing standardized guidelines for banking practices and uniform quality assessment measures (Chapter 2). Comparing current forensic and genomic markers revealed similar concordance and discordance rates with a slight performance advantage towards the forensic markers. The results indicate that multiple runs are necessary to ensure reliability (Chapter 3). A significant ethical issue arises from the forensic practice of surreptitious DNA sampling. This lack of transparency violates autonomy, threatens the legitimacy of the State's int / L'analyse moléculaire des variations de l'ADN a supplanté toutes les technologies médicolégales antérieures d'identification des victimes et des suspects. Bien que le champ de l'éthique ait tenté de gérer la pléthore d'information génétique disponible, particulièrement dans les applications cliniques, il y a eu peu d'examen des enjeux éthiques émergeants dans le domaine médicolégal. La recherche juridique met en évidence certains aspects des enjeux émergeant avec une pertinence particulière pour les défis auxquels les tribunaux sont confrontés ainsi que les défis à l'égard des libertés individuelles.L'objectif général de cette thèse était d'évaluer la validité scientifique, l'acceptabilité éthique et la responsabilité légale dans les applications médicolégales de la génétique moléculaire. En particulier, la science contemporaine nous a permis d'accéder à des informations qui vont au-delà de ce qui était anticipé à l'origine si bien que des traces d'ADN peuvent être obtenues trivialement de tout individu. En conséquence, l'étendue et la composition des banques existantes d'ADN excèdent de loin le mandat législatif. Le premier chapitre revoit les standards légaux d'évidence et évalue le niveau d'exactitude nécessaire afin que les tests d'ADN médico-légaux rencontrent les standards d'évidence. Une évaluation des pratiques actuelles dans la mise en banque d'ADN a révélé des pratiques de consentement éclairé adéquate, le besoin de réexaminer l'accès aux échantillons de santé publique en portant l'attention aux intérêts des populations locales et la nécessité de développer des lignes directrices standardisées pour les pratiques de mise en banque et de mesures uniformes de l'évaluation de la qualité (chapitre 2). La comparaison des marqueurs médicolégaux actuels aux marqueurs génomiques a révélé des taux de concordance et de discord

The urogenital system performs essential functions for life and reproduction in all vertebrate species. In humans, congenital malformations of the urogenital system rank among the most common birth defects and can cause life-threatening complications in adults. Therefore, understanding urogenital system pathogenesis is imperative in order to provide more accurate diagnoses and better medical interventions to ultimately improve clinical outcome in patients. In this regard, my thesis research projects focused on the development of the mouse urogenital system with the aim of advancing our current knowledge of the biological processes underlying both its normal morphogenesis and disease. Pax transcription factors are critical regulatory proteins involved in the development of many organs, including the urogenital system. Hence, the results presented here, address the role of Pax2/8 along with one of their important effectors, Gata3, at different stages of urogenital system morphogenesis. In the definitive (metanephric) kidney, Pax2/8 regulate both kidney growth and nephron differentiation. Interestingly, from a molecular perspective, this is partly achieved by Pax2/8 inducing the transcription of potent target genes, such as Wnt11 and Lim1, as well as the positive effect of Pax genes on nephric cell survival. In the embryonic (pro/mesonephric) kidney, Pax2/8 activate the expression of the transcription factor Gata3, which in turn controls several important cellular processes, like cell proliferation, morphogenesis, guidance and differentiation. Importantly, as a downstream mediator of Wnt/beta-catenin signaling, Gata3 assures proper metanephric kidney induction by preventing an excessive cellular response to local growth factors. The search for Gata3-regulated genes in the mesonephros identified the Ret receptor tyrosine kinase and the transcription factor Lim1, as critical downstream effectors and suggested / Le système urogénital accomplit des fonctions essentielles pour la vie et la reproduction de tous les vertébrés. Chez l'humain, les malformations du système urogénital sont parmi les plus fréquentes anomalies congénitales et peuvent entraîner de graves complications chez l'adulte. Afin d'améliorer le diagnostique, les interventions médicales ainsi que le pronostique chez les patients, une meilleure connaissance de la pathogénèse de l'appareil urinaire et génital est nécessaire. Ainsi, l'objectif de mon programme de recherche de thèse s'est basé sur l'étude du développement embryonnaire du système urogénital chez la souris, dans le but d'améliorer notre compréhension des processus biologiques sous-jacents à la morphogénèse ainsi qu'à la pathogénèse rénale.Les facteurs de transcription Pax sont des protéines qui régulent le développement de plusieurs organes, tels que ceux du système urogénital. Les résultats de recherche présentés ici, traitent des fonctions de Pax2/8 ainsi qu'un de leur effecteur principal Gata3, à différents stades de développement du système génito-urinaire. Pax2/8 régulent la croissance du rein définitif (métanéphros) et la différentiation des néphrons. D'un point de vue moléculaire, ceci est en partie dû à l'activation de la transcription de gênes cibles comme Wnt11 et Lim1, ainsi qu'à l'effet positif exercé par les gênes Pax sur la survie cellulaire de l'appareil rénal. Dans le rein embryonnaire (pro/mésonephros), Pax2/8 activent l'expression du facteur de transcription Gata3, qui lui-même contrôle plusieurs processus cellulaires, comme la prolifération, la morphogénèse, le guidage et la différentiation. Notablement, en tant que médiateur en aval de la signalisation Wnt/beta-catenin, Gata3 assure une induction métanéphrique adéquate en empêchant une réponse cellulaire excessive aux facteurs de croissance locaux. L

Genomic imprinting is a mechanism of fine regulation of gene expression. Imprinted genes are expressed from only one parental allele and many of them have critical roles in growth and development. Imprinting marks that distinguish the parental origin must be erased and re-established in germ cells according to the sex of the individual to ensure proper embryonic development. We investigated occurrence of imprinting errors in mouse embryos. Firstly, we tested the hypothesis that errors in resetting of imprints occur and lead to grandparental-origin effects on embryonic growth. Although we did not find statistically significant effects, we observed trends that should be confirmed by replication. Secondly, we examined expression of 5 genes located on the distal part of chromosome 12 in order to establish the incidence of spontaneous imprinting errors. We report a strain and parental-origin specific imprinting relaxation of Dlk1 and Dio3 genes.

The signaling peptide Indian Hedgehog (Ihh) is secreted by prehypertrophic chondrocytes of the growth plate, the growth center of long bones. It functions to negatively regulate differentiation and positively regulate proliferation of chondrocytes. Thus, Ihh ultimately controls the rate of bone growth. / The transcriptional regulation of the Ihh gene, however, remains uncharacterized. In order to study the gene's regulation, the genomic Ihh sequences from several species were aligned to identify conserved regions that may contain regulatory sites. Two putative Stat transcription factor binding sites were identified, one of which is conserved across all species studied while the other is rodent-specific. / In addition, an in vitro system was established to test the upstream region of the gene for transcriptional activity. ATDC5 chondrogenic cells were stably transfected with a plasmid containing 5kb of sequence located upstream of Ihh as well as a luciferase reporter gene. The presence of the Ihh sequence induced expression of the luciferase reporter 50 fold above expression from a control plasmid. COS7 and ATDC5 cells transiently transfected with similar Ihh-luciferase constructs resulted in unique induction patterns. Thus, the Ihh upstream genomic region contains sequences that regulate expression in a tissue-specific fashion.

The majority of genetic traits including most common diseases are believed to be multigenic and arise both from variations in coding sequences as well as from regulatory polymorphisms. Genome-wide approaches are needed to develop models for understanding this complexity. This thesis develops approaches for studying genetic variation affecting gene expression on a genome-wide scale. This included development of experimental design principles and analytical methods for microarray data. These principles were then applied to characterize differences between commonly-used A/J and C57BL/6J inbred mouse strains at the molecular level identifying over 2000 genes differentially expressed between these strains across 4 tissues. To further investigate the role of genetic variation in genome-wide expression changes, we analyzed expression profiles of lung tissue obtained from a panel of recombinant congenic strains (RCS) derived from the same inbred strains. An ANOVA was applied using a model to test the association of expression profiles with donor-strain of origin (DSO, inferred from RCS genotyping data), and with genetic background. This model identified over 1500 genes whose expression levels were associated with DSO status (P<0.05) having adjusted for the variability due to predominant strain of background, suggestive of cis-regulatory variation in these genes. We randomly selected 50 positive genes displaying association between DSO and 80 negative genes for validation using allelic imbalance (AI), a method that uses intragenic SNPs for detecting genes with cis-regulatory variation that measures allele-specific transcript levels in cDNA of heterozygous individuals. Of the genes chosen, 54% of positive versus 27% of negative genes contained at least one SNP within ≥ 1 kbp of 3' UTR sequenced (P<0.05 Fisher exact test). Al was found in 63% of positive genes versus 23% of negative genes (P<0.01 Fisher exact test) representing a greater than 10-fo

A total of 95,678,311 feed records from January 2000 to May 2007, corresponding to 16,866,117 test-day records or 1,714,651 cow lactation records obtained from Quebec Dairy Herd Improvement program, VALACTA were used to estimate heritabilities of feed intake. Genetic parameters of feed intake were estimated using both a single and two-trait (with milk yield as the second trait) animal models which included herd-year-season (hys) at calving, age at calving, male and female phantom groups as fixed effects, and animal and residual as random effects (i.e. for analysis of complete lactation feed intake in parity 1, there were 119,620 cows from 1,248 sires and 88,500 dams, 20,133 levels of hys, 20 levels of age at calving and 49 phantom groups). The model was fitted by Restricted Maximum Likelihood and relationships among animals were taken into account on both the male and female side of the pedigree (i.e. for analysis of complete lactation feed intake in parity 1, there were 308,029 animals from 6 generations in the pedigree file). Heritability estimates obtained from the single trait model ranged from 0.04 to 0.14 for complete lactation feed intake traits, 0.01 to 0.03 for 90-d feed intake traits, 0.08 to 0.19 for 305-d feed intake traits, and 0.10 to 0.21 for daily feed intake traits. Heritability estimates for 305-d dry matter intake obtained from the two-trait model were 0.06, 0.01 and 0.08 in parity 1, 2 and 3 respectively. / Un total de 95 678 311 relevés de consommation de nourriture récoltés entre janvier 2000 et mai 2007, correspondant à 16 866 117 relevés Jour de Test ou à 1 714 651 relevés de lactation de vaches, ont été obtenus grâce au programme d'analyse des troupeaux laitiers du Québec, VALACTA, et été utilisés afin d'estimer l'héritabilité de la consommation. Les paramètres génétiques de consommation ont été établis en utilisant des modèles d'animaux à un ou deux caractères. Bien que le second caractère était toujours la production de lait, les autres caractères utilisés incluaient des effets fixes tels que des effets troupeau-année-saison au vêlage, âge au vêlage, les groupes d'ascendance paternelle et maternelle, ainsi que des effets aléatoires tels qu'animal et résiduels. Par exemple, pour l'analyse de la comsommation complète pendant la lactation de la parité 1: il y avait 119 620 vaches issues de 1 248 taureaux et 88 500 vaches, 20 133 niveaux de troupeau-année-saison, 20 niveaux d'âge au moment d u vêlage, et 49 groupes d'ascendance paternelle et maternelle. Les modèles ont été formatés selon une méthode de vraisemblance maximale restreinte (REML) et en prenant en compte les liens génétiques entre animaux, lesquels ont été inclus dans les pedigrées des males et des femelles. Ainsi, pour l'analyse de la comsommation complète pendant la lactation de la parité 1, il y avait 308 029 animaux de 6 générations dans le fichier pedigrée. Les estimations de l' héritabilité obtenues à l'issu du modèle à un caractère variaient de 0.04 à 0.14 pour la consommation complète pendant la lactation, de 0.01 à 0.03 pour les caratères obtenus suite à une consommation sur 90 jours, de 0.08 a 0.19 suite à une consommation de 305 jours, et de 0.10 à 0.21 pour les caractères de consommation journalière. L'héritabilité estimée depuis le modèle à deux caratères et suite à 305 jours de consommation de matière sèche é

Lipo-chitooligosaccharides (LCOs) produced by N2-fixing rhizobacteria initiate host nodule formation. Foliar application of LCOs has been shown to induce stress-related genes under optimal growth conditions. To study the effects of LCO foliar spray under stressed conditions, soybean seedlings grown at optimal temperature were exposed to sub-optimal temperature. After a 5-day acclimation period, the first trifoliolate leaves were sprayed with 10-7 M LCO (NodBj-V (C18:1, MeFuc)) produced by Bradyrhizobium japonicum, and harvested at 0 and 48 h following treatment. Microarray analysis was performed using Affymetrix GeneChip® Soybean Genome Arrays. A total of 147 genes were differentially expressed 48 h after LCO treatment, including a number of stress-related genes and transcription factors. In addition, during the 48 h following treatment, hundreds of genes were differentially expressed in LCO-treated plants, indicating that the dynamic soybean foliar transcriptome was highly responsive to LCO treatment. The microarray data was supported by quantitative real-time PCR data. / Les lipo-chitooligosaccharides (LCOs) produits par les rhizobactéries initient la formation de nodules chez l'hôte. La vaporisation foliaire de LCOs provoque l'expression des gènes liés aux stress et à la transduction du signal. Pour étudier les effets de la vaporisation foliaire de LCOs en condition stressante, des plants de soja qui étaient cultivés à température optimale ont été exposés à température sub-optimale. Après une période d'acclimatation de 5 jours, les premiers feuillages ont été vaporisés de 10-7 M LCO (NodBj-V (C18:1, MeFuc)) produit par Bradyrhizobium japonicum, et ont été recueillis 0 h et 48 h après traitement. L'analyse microarray a été réalisée par Affymetrix GeneChip® Soybean Genome Arrays. Un total de 147 gènes se sont exprimés différemment 48 h après application du traitement LCO, y compris les gènes liés aux stress et à la transduction du signal. De plus, durant les 48 h suivant la vaporisation foliaire, des centaines de gènes qui avaient montré une expression différentielle étaient spécifiques aux plantes traitées avec LCO. Ces résultats indiquent que le transcriptome foliaire dynamique du soja a réagi au traitement LCO. Les données de microarray étaient supportées par des données de PCR quantitatif en temps réel.

Hydatidiform mole is a form of gestational trophoblastic disease characterized by absence of embryo and hydropic degeneration of the chorionic villi. The majority of complete hydatidiform moles are sporadic and androgenetic; however, a rare subset has been identified which has a biparental genome. These have been found occasionally to recur in the same woman and, in multiple women in the same family suggesting a genetic defect. A maternal locus for biparental familial moles has been mapped to a 1.1Mb region on 19q13.4. Genotyping of a new family with multiple affected sisters has lead to the narrowing of this interval to 0.65Mb, which contains 30 genes. Screening of these genes led to the discovery of mutations in NALP7 in all affected individuals from four families of different ethnic origin. NALP7 is a member of a family of cytoplasmic proteins known to play a role in inflammation and the innate immune system.

Cystic Fibrosis (CF) is the most common autosomal recessive disease affecting the Caucasian population. This devastating disease is caused by any one of 1500 mutations identified in the Cystic Fibrosis Transmembrane Regulator Conductance (cftr) gene. Chronic inflammation is a hallmark of CF and it affects all systems including respiratory, gastrointestinal, reproductive and skeletal. Although the exact molecular link between the CFTR dysfunction and various phenotypes remains to be delineated, many phenotypes seem to be linked to inadequate nutritional absorption of essential fatty acids and vitamins, which leads to an imbalance between the essential fatty acids docosahexaenoic acid (DHA) and arachidonic acid (AA). The skeletal system does not only serve as mechanical support, but also functions as an active organ that regulates balance and interactions between both local and systemic hormones, cytokines and prostaglandins. Previously our laboratory has shown that fenretinide [ N-(4-hydroxyphenyl) retinamide] corrects the essential fatty acid imbalance. We hypothesized that correcting the DHA/AA ratio in the plasma of Cftr-KO mice could avoid the early-onset osteoporosis. This thesis presents our novel results describing how fenretinide prevents osteoporosis. We found that twice a week treatment with fenretinide over a period of four weeks dramatically increased trabecular bone volume and quality in Cftr-KO mice. The results of this thesis strongly suggest that fenretinide might have potential for the treatment of cystic fibrosis patients by preventing the reduction of trabecular bone mineral density.